Author Contributions
Conceptualization, E.C. and J.P.; Data curation, E.C., D.M. and E.R.; Formal analysis, J.P.; Funding acquisition, J.P.; Investigation, E.C. and J.P.; Methodology, E.C. and D.M.; Software, E.C.; Supervision, J.P.; Validation, E.C., D.M. and E.R.; Visualization, E.C.; Writing—original draft, D.M.; Writing—review & editing, J.P. All authors have read and agreed to the published version of the manuscript.
Figure 1.
(a) APR-03 mobile robot operated by an author of this paper. (b) Close-up image of the onboard LIDAR sensor. (c) Detail of the onboard LIDAR without the protective housing cover.
Figure 1.
(a) APR-03 mobile robot operated by an author of this paper. (b) Close-up image of the onboard LIDAR sensor. (c) Detail of the onboard LIDAR without the protective housing cover.
Figure 2.
Reference point cloud map of the experimentation facility.
Figure 2.
Reference point cloud map of the experimentation facility.
Figure 3.
Top (up) and lateral (down) view of different scans obtained when placing a small box on the ground at approximately 1000 mm from the LIDAR. The height of the LIDAR was 380 mm and the tilt was changed from 45° (reddish ground line, close to X = 0) to 80° (violet ground line) in increments of 5°.
Figure 3.
Top (up) and lateral (down) view of different scans obtained when placing a small box on the ground at approximately 1000 mm from the LIDAR. The height of the LIDAR was 380 mm and the tilt was changed from 45° (reddish ground line, close to X = 0) to 80° (violet ground line) in increments of 5°.
Figure 4.
Top (up) and lateral (down) view of the different scans obtained when placing one shoe/leg at a fixed distance from the LIDAR. The tilt of the LIDAR has been changed from 0° (horizontal) to 45° in increments of 5°.
Figure 4.
Top (up) and lateral (down) view of the different scans obtained when placing one shoe/leg at a fixed distance from the LIDAR. The tilt of the LIDAR has been changed from 0° (horizontal) to 45° in increments of 5°.
Figure 5.
Top (up) and lateral (down) view of the different scans obtained when having stairs going down at a fixed distance of 1250 mm depending on the 2D LIDAR tilt angle. The tilt of the LIDAR has been changed from 0° (horizontal) to 45° in increments of 5°. The distance points originated by the surrounding walls difficult the visual interpretation of the data shown.
Figure 5.
Top (up) and lateral (down) view of the different scans obtained when having stairs going down at a fixed distance of 1250 mm depending on the 2D LIDAR tilt angle. The tilt of the LIDAR has been changed from 0° (horizontal) to 45° in increments of 5°. The distance points originated by the surrounding walls difficult the visual interpretation of the data shown.
Figure 6.
Top (up) and lateral (down) view of the different scans obtained when having stairs going up at a fixed distance of 750 mm in front of the 2D LIDAR depending on tilt angle. The tilt of the LIDAR has been changed from 0° (horizontal) to 45° in increments of 5°. The distance points originated by the surrounding walls have been eliminated for enhanced visual interpretation.
Figure 6.
Top (up) and lateral (down) view of the different scans obtained when having stairs going up at a fixed distance of 750 mm in front of the 2D LIDAR depending on tilt angle. The tilt of the LIDAR has been changed from 0° (horizontal) to 45° in increments of 5°. The distance points originated by the surrounding walls have been eliminated for enhanced visual interpretation.
Figure 7.
Mobile robot APR-03 placed in front of: (a) a stair going down and (b) a stair going up.
Figure 7.
Mobile robot APR-03 placed in front of: (a) a stair going down and (b) a stair going up.
Figure 8.
Experiment in front of a stairs going down: (a) raw scan data, (b) 3D representation. Distance points classified as ground (green), expected location of the ground (dotted red line).
Figure 8.
Experiment in front of a stairs going down: (a) raw scan data, (b) 3D representation. Distance points classified as ground (green), expected location of the ground (dotted red line).
Figure 9.
Experiment in front of a stairs going up: (a) raw scan data, (b) 3D representation.
Figure 9.
Experiment in front of a stairs going up: (a) raw scan data, (b) 3D representation.
Figure 10.
Illustration of the strategy applied to tilt down 25° the onboard 2D LIDAR. The horizontal line depicts the location of the measurement plane of the laser beam.
Figure 10.
Illustration of the strategy applied to tilt down 25° the onboard 2D LIDAR. The horizontal line depicts the location of the measurement plane of the laser beam.
Figure 11.
Original 2D scanned point cloud (blue) and its 2D projection in the horizontal plane of the reference map (green). The lines depicts the correspondence between the original and projected scan points: (light blue) cases over the ground level and (magenta) cases under the ground level.
Figure 11.
Original 2D scanned point cloud (blue) and its 2D projection in the horizontal plane of the reference map (green). The lines depicts the correspondence between the original and projected scan points: (light blue) cases over the ground level and (magenta) cases under the ground level.
Figure 12.
Mobile robot APR-03 in the corridor: (a) at the right side and (b) at the left side.
Figure 12.
Mobile robot APR-03 in the corridor: (a) at the right side and (b) at the left side.
Figure 13.
2D projection of some scans gathered by the LIDAR placed at the starting positions shown in
Figure 12: (top) right side of the corridor and (down) left side of the corridor.
Figure 13.
2D projection of some scans gathered by the LIDAR placed at the starting positions shown in
Figure 12: (top) right side of the corridor and (down) left side of the corridor.
Figure 14.
Estimated trajectory of the mobile robot with the 2D LIDAR tilted down 25° for the cases shown in
Figure 12: (top) right side of the corridor and (down) left side of the corridor.
Figure 14.
Estimated trajectory of the mobile robot with the 2D LIDAR tilted down 25° for the cases shown in
Figure 12: (top) right side of the corridor and (down) left side of the corridor.
Figure 15.
Image of the mobile robot APR-03 at the stair area of the facility.
Figure 15.
Image of the mobile robot APR-03 at the stair area of the facility.
Figure 16.
2D projection of a scan gathered by the LIDAR placed at the starting position shown in
Figure 15. The arrow depicts the location and orientation of the mobile robot, the green dots depict scan points classified as ground points.
Figure 16.
2D projection of a scan gathered by the LIDAR placed at the starting position shown in
Figure 15. The arrow depicts the location and orientation of the mobile robot, the green dots depict scan points classified as ground points.
Figure 17.
Estimated trajectory of the mobile robot with the 2D LIDAR tilted down 25° in one experiment performed in the stairs area of the facility.
Figure 17.
Estimated trajectory of the mobile robot with the 2D LIDAR tilted down 25° in one experiment performed in the stairs area of the facility.
Table 1.
Illustrative images of the validation detection experiments: small box, one shoe/leg, stair going down, and stairs going up.
Table 2.
Small box detection results obtained for different distances and LIDAR tilt angles: (√) object detected, (R) object detected after ground reflection, (-) object not detected.
Table 2.
Small box detection results obtained for different distances and LIDAR tilt angles: (√) object detected, (R) object detected after ground reflection, (-) object not detected.
| Tilt |
---|
Distance | 0° | 5° | 10° | 15° | 20° | 25° | 30° | 35° | 40° | 45° |
---|
250 mm | - | - | - | - | - | - | - | - | √ | √ |
500 mm | - | - | - | - | - | √ | √ | √ | - | - |
750 mm | - | - | - | - | √ | √ | - | - | - | - |
1000 mm | - | - | - | √ | √ | - | - | - | - | - |
1250 mm | - | - | - | √ | - | - | - | - | - | - |
1500 mm | - | - | √ | √ | - | - | - | - | - | - |
1750 mm | - | - | √ | - | - | - | - | - | - | - |
2000 mm | - | - | √ | - | - | - | - | - | - | - |
2250 mm | - | - | R | - | - | - | - | - | - | - |
2500 mm | - | - | R | - | - | - | - | - | - | - |
2750 mm | - | - | - | - | - | - | - | - | - | - |
3000 mm | - | √ | - | - | - | - | - | - | - | - |
3250 mm | - | √ | - | - | - | - | - | - | - | - |
3500 mm | - | √ | - | - | - | - | - | - | - | - |
3750 mm | - | √ | - | - | - | - | - | - | - | - |
4000 mm | - | √ | - | - | - | - | - | - | - | - |
Table 3.
One shoe/leg detection results obtained for different distance and LIDAR tilt angles: (√) object detected, (R) object detected after ground reflection, (-) object not detected.
Table 3.
One shoe/leg detection results obtained for different distance and LIDAR tilt angles: (√) object detected, (R) object detected after ground reflection, (-) object not detected.
| Tilt |
---|
Distance | 0° | 5° | 10° | 15° | 20° | 25° | 30° | 35° | 40° | 45° |
---|
250 mm | √ | √ | √ | √ | √ | √ | √ | √ | √ | √ |
500 mm | √ | √ | √ | √ | √ | √ | √ | √ | - | - |
750 mm | √ | √ | √ | √ | √ | √ | - | - | - | - |
1000 mm | √ | √ | √ | √ | √ | - | - | - | - | - |
1250 mm | √ | √ | √ | √ | - | - | - | - | - | - |
1500 mm | √ | √ | √ | - | - | - | - | - | - | - |
1750 mm | √ | √ | √ | - | - | - | - | - | - | - |
2000 mm | √ | √ | √ | - | - | - | - | - | - | - |
2250 mm | √ | √ | R | - | - | - | - | - | - | - |
2500 mm | √ | √ | R | - | - | - | - | - | - | - |
2750 mm | √ | √ | R | - | - | - | - | - | - | - |
3000 mm | √ | √ | R | - | - | - | - | - | - | - |
3250 mm | √ | √ | R | - | - | - | - | - | - | - |
3500 mm | √ | √ | R | - | - | - | - | - | - | - |
3750 mm | √ | √ | - | - | - | - | - | - | - | - |
4000 mm | √ | √ | - | - | - | - | - | - | - | - |
Table 4.
Stairs going down detection results obtained for different distances and LIDAR tilt angles: (S) hole or stair going down detected, (G) ground detected at the expected location, (numeric value) average distance of a frontal obstacle.
Table 4.
Stairs going down detection results obtained for different distances and LIDAR tilt angles: (S) hole or stair going down detected, (G) ground detected at the expected location, (numeric value) average distance of a frontal obstacle.
Down Stairs Distance | Tilt |
---|
0° | 5° | 10° | 15° | 20° | 25° | 30° | 35° | 40° | 45° |
---|
250 mm | 5632 mm | S | S | S | S | S | S | S | S | S |
500 mm | 5856 mm | S | S | S | S | S | S | S | G | G |
750 mm | 6071 mm | S | S | S | S | S | G | G | G | G |
1000 mm | 6340 mm | S | S | S | S | G | G | G | G | G |
1250 mm | 6642 mm | S | S | S | G | G | G | G | G | G |
1500 mm | 6907 mm | S | S | G | G | G | G | G | G | G |
Table 5.
Stairs going up detection results obtained for different distances and LIDAR tilt angles: (G) ground detected at its expected location, (numeric value) average distance in mm of the frontal obstacle detected.
Table 5.
Stairs going up detection results obtained for different distances and LIDAR tilt angles: (G) ground detected at its expected location, (numeric value) average distance in mm of the frontal obstacle detected.
Up Stairs Distance | Tilt |
---|
0° | 5° | 10° | 15° | 20° | 25° | 30° | 35° | 40° | 45° |
---|
250 mm | 780 | 492 | 510 | 499 | 515 | 430 | 350 | 270 | 265 | 251 |
500 mm | 1012 | 729 | 764 | 759 | 575 | 500 | 479 | 500 | G | G |
750 mm | 1305 | 1036 | 1010 | 787 | 739 | 765 | G | G | G | G |
1000 mm | 1596 | 1291 | 1254 | 984 | 1001 | G | G | G | G | G |
1250 mm | 1812 | 1513 | 1267 | 1239 | G | G | G | G | G | G |
1500 mm | 2042 | 1746 | 1151 | G | G | G | G | G | G | G |